DESCRIPTION: Detection and quantitation of nucleic acid sequences bearing rare cancer driver mutations is currently used to inform clinical treatment decisions such as use or discontinuation or therapeutics. DNA diagnostics can also be used to perform noninvasive early-stage cancer screening based on cell-free nucleic acids, but requires the development of an inexpensive, multiplex, and mutation-sensitive platform. Current methods to detect rare mutations exhibit either low mutation sensitivity (NGS) or poor multiplexing (digital PCR), and are frequently expensive and/or labor-intensive (both NGS and digital PCR). The PI proposes to develop a PCR platform capable of simultaneous analysis of 1000 point mutations at 0.1% single-base mutation sensitivity from a single 10 ng DNA sample. The polymerase chain reaction (PCR) remains most robust, simple, sensitive, and popular method for DNA diagnostics. Allele-speci c PCR primers can discriminate single nucleotide variants (SNVs) at roughly 1 part in 100, but it is dicult to design primers that can function in a highly (>10) multiplexed setting while preserving high SNV selectivity. NGS has been adopted by researchers and clinicians for highly multiplexed genetic analysis of cancer tissue samples, but it is not practical to obtain sucient read depth to allow <0.1% mutation sen- sitivity. Imperfect solutions to the multiplexing limitation of PCR include (1) sample splitting, which requires signi cantly larger sample input quantities to preserve clinical sensitivity, and (2) nested PCR/pre-ampli cation which necessitates complex equipment or an open-tube step that risks contamination. The two primary challenges for highly multiplexed allele-speci c PCR are (1) temperature robustness and (2) primer-dimer suppression. Allele-speci c PCR primers are capable of rare allele detection/ampli cation only within a narrow window of about 1 C, the exact value of which cannot be accurately predicted from sequence. The PI is developing novel primers that are sensitive to single-base mutations at 0.1% allele frequency across a window of 8 C; this broad temperature robustness overcomes the diculty of empirically optimizing many primers to have the same melting temperature. PCR primers in general can spuriously hybridize to each other and form primer-dimers that result in nonsense ampli cation products; primer design grows sharply more dicult as the number of primers simultaneously in solution increases. The PI is developing novel primers that are mostly double-stranded and thus signi cantly less likely to result in primer-dimer interactions; the suppression of nonspeci c primer hybridization enables PCR with signi cantly higher multiplexing. Finally, the research team will develop a novel on-chip convection PCR instrument with an integrated mul- tiplexed array-based readout. Importantly, no open-tube transfer of PCR amplicons is necessary, preventing contamination that leads to false positive. Signi cant preliminary data has been obtained for all three aspects o the proposed research project, and the research team plans to apply the developed technology initially to retrospective analysis of non-small cell lung cancer (NSCLC) patient blood samples.